Classification of High-Spatial-Resolution Remote Sensing Scenes Method Using Transfer Learning and Deep Convolutional Neural Network
The deep convolutional neural network (DeCNN) is considered one of promising techniques for classifying the high-spatial-resolution remote sensing (HSRRS) scenes, due to its powerful feature extraction capabilities. It is well-known that huge high-quality labeled datasets are required for achieving...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9085918/ |
| _version_ | 1828383494895566848 |
|---|---|
| author | Wenmei Li Ziteng Wang Yu Wang Jiaqi Wu Juan Wang Yan Jia Guan Gui |
| author_facet | Wenmei Li Ziteng Wang Yu Wang Jiaqi Wu Juan Wang Yan Jia Guan Gui |
| author_sort | Wenmei Li |
| collection | DOAJ |
| description | The deep convolutional neural network (DeCNN) is considered one of promising techniques for classifying the high-spatial-resolution remote sensing (HSRRS) scenes, due to its powerful feature extraction capabilities. It is well-known that huge high-quality labeled datasets are required for achieving the better classification performances and preventing overfitting, during the training DeCNN model process. However, the lack of high-quality datasets limits the applications of DeCNN. In order to solve this problem, in this article, we propose a HSRRS image scene classification method using transfer learning and the DeCNN (TL-DeCNN) model in a few shot HSRRS scene samples. Specifically, three typical DeCNNs of VGG19, ResNet50, and InceptionV3, trained on the ImageNet2015, the weights of their convolutional layer for that of the TL-DeCNN are transferred, respectively. Then, TL-DeCNN just needs to fine-tune its classification module on the few shot HSRRS scene samples in a few epochs. Experimental results indicate that our proposed TL-DeCNN method provides absolute dominance results without overfitting, when compared with the VGG19, ResNet50, and InceptionV3, directly trained on the few shot samples. |
| first_indexed | 2024-12-10T04:50:16Z |
| format | Article |
| id | doaj.art-82fac23080f54746955633260c5d2022 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-10T04:50:16Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-82fac23080f54746955633260c5d20222022-12-22T02:01:38ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352020-01-01131986199510.1109/JSTARS.2020.29884779085918Classification of High-Spatial-Resolution Remote Sensing Scenes Method Using Transfer Learning and Deep Convolutional Neural NetworkWenmei Li0https://orcid.org/0000-0002-1108-0507Ziteng Wang1Yu Wang2https://orcid.org/0000-0001-7763-4261Jiaqi Wu3https://orcid.org/0000-0002-3299-7410Juan Wang4https://orcid.org/0000-0003-4291-177XYan Jia5Guan Gui6https://orcid.org/0000-0003-3888-2881School of Geographic and Biologic Information, Nanjing University of Posts and Telecommunications, Nanjing, ChinaCollege of Telecommunications and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing, ChinaCollege of Telecommunications and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing, ChinaSchool of Geographic and Biologic Information, Nanjing University of Posts and Telecommunications, Nanjing, ChinaCollege of Telecommunications and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing, ChinaSchool of Geographic and Biologic Information, Nanjing University of Posts and Telecommunications, Nanjing, ChinaCollege of Telecommunications and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing, ChinaThe deep convolutional neural network (DeCNN) is considered one of promising techniques for classifying the high-spatial-resolution remote sensing (HSRRS) scenes, due to its powerful feature extraction capabilities. It is well-known that huge high-quality labeled datasets are required for achieving the better classification performances and preventing overfitting, during the training DeCNN model process. However, the lack of high-quality datasets limits the applications of DeCNN. In order to solve this problem, in this article, we propose a HSRRS image scene classification method using transfer learning and the DeCNN (TL-DeCNN) model in a few shot HSRRS scene samples. Specifically, three typical DeCNNs of VGG19, ResNet50, and InceptionV3, trained on the ImageNet2015, the weights of their convolutional layer for that of the TL-DeCNN are transferred, respectively. Then, TL-DeCNN just needs to fine-tune its classification module on the few shot HSRRS scene samples in a few epochs. Experimental results indicate that our proposed TL-DeCNN method provides absolute dominance results without overfitting, when compared with the VGG19, ResNet50, and InceptionV3, directly trained on the few shot samples.https://ieeexplore.ieee.org/document/9085918/Deep convolutional neural network (DeCNN)few shothigh-spatial-resolution remote sensing (HSRRS)scene classificationtransfer learning |
| spellingShingle | Wenmei Li Ziteng Wang Yu Wang Jiaqi Wu Juan Wang Yan Jia Guan Gui Classification of High-Spatial-Resolution Remote Sensing Scenes Method Using Transfer Learning and Deep Convolutional Neural Network IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep convolutional neural network (DeCNN) few shot high-spatial-resolution remote sensing (HSRRS) scene classification transfer learning |
| title | Classification of High-Spatial-Resolution Remote Sensing Scenes Method Using Transfer Learning and Deep Convolutional Neural Network |
| title_full | Classification of High-Spatial-Resolution Remote Sensing Scenes Method Using Transfer Learning and Deep Convolutional Neural Network |
| title_fullStr | Classification of High-Spatial-Resolution Remote Sensing Scenes Method Using Transfer Learning and Deep Convolutional Neural Network |
| title_full_unstemmed | Classification of High-Spatial-Resolution Remote Sensing Scenes Method Using Transfer Learning and Deep Convolutional Neural Network |
| title_short | Classification of High-Spatial-Resolution Remote Sensing Scenes Method Using Transfer Learning and Deep Convolutional Neural Network |
| title_sort | classification of high spatial resolution remote sensing scenes method using transfer learning and deep convolutional neural network |
| topic | Deep convolutional neural network (DeCNN) few shot high-spatial-resolution remote sensing (HSRRS) scene classification transfer learning |
| url | https://ieeexplore.ieee.org/document/9085918/ |
| work_keys_str_mv | AT wenmeili classificationofhighspatialresolutionremotesensingscenesmethodusingtransferlearninganddeepconvolutionalneuralnetwork AT zitengwang classificationofhighspatialresolutionremotesensingscenesmethodusingtransferlearninganddeepconvolutionalneuralnetwork AT yuwang classificationofhighspatialresolutionremotesensingscenesmethodusingtransferlearninganddeepconvolutionalneuralnetwork AT jiaqiwu classificationofhighspatialresolutionremotesensingscenesmethodusingtransferlearninganddeepconvolutionalneuralnetwork AT juanwang classificationofhighspatialresolutionremotesensingscenesmethodusingtransferlearninganddeepconvolutionalneuralnetwork AT yanjia classificationofhighspatialresolutionremotesensingscenesmethodusingtransferlearninganddeepconvolutionalneuralnetwork AT guangui classificationofhighspatialresolutionremotesensingscenesmethodusingtransferlearninganddeepconvolutionalneuralnetwork |